检索结果-内蒙古大学图书馆

26th IEEE International Conference on Software Analysis, Evolution and Reengineering (SANER)

作者： Bui, Nghi D. Q. Yu, Yijun Jiang, Lingxiao Singapore Management Univ Sch Informat Syst Singapore Singapore Open Univ Ctr Res Comp Milton Keynes Bucks England

ISBN: (纸本)9781728105918

algorithm classification is to automatically identify the classes of a program based on the algorithm(s) and/or data structure(s) implemented in the program. It can be useful for various tasks, such as code reuse, code theft detection, and malware detection. Code similarity metrics, on the basis of features extracted from syntax and semantics, have been used to classify programs. Such features, however, often need manual selection effort and are specific to individual programming languages, limiting the classifiers to programs in the same language. To recognize the similarities and differences among algorithms implemented in different languages, this paper describes a framework of Bilateral Neural Networks (Bi-NN) that builds a neural network on top of two underlying sub-networks, each of which encodes syntax and semantics of code in one language. A whole Bi-NN can be trained with bilateral programs that implement the same algorithms and/or data structures in different languages and then be applied to recognize algorithm classes across languages. We have instantiated the framework with several kinds of token-, tree-and graph-based neural networks that encode and learn various kinds of information in code. We have applied the instances of the framework to a code corpus collected from GitHub containing thousands of Java and C++ programs implementing 50 different algorithms and data structures. Our evaluation results show that the use of Bi-NN indeed produces promising algorithm classification results both within one language and across languages, and the encoding of dependencies from code into the underlying neural networks helps improve algorithm classification accuracy further. In particular, our custom-built dependency trees with tree-based convolutional neural networks achieve the highest classification accuracy among the different instances of the framework that we have evaluated. Our study points to a possible future research direction to tailor bilateral and mu

关键词： cross-language mapping program classification algorithm classification code embedding code dependency neural network bilateral neural network

来源：评论

学校读者我要写书评

暂无评论

Malware algorithm classification method based on big data analysis

引用

INTERNATIONAL JOURNAL OF WEB AND GRID SERVICES 2017年第1期13卷 112-130页

作者： Zhao, Jingling Chen, Shilei Cao, Mengchen Cui, Baojiang Beijing Univ Posts & Telecommun Sch Comp Sci Natl Engn Lab Mobile Network Secur Beijing 100876 Peoples R China

Internet technology has greatly increased the number of malware attacks on networks. Consequently, it has also elevated the importance of automatic malware detection and classification technology based on big data analysis in the field of information security. This paper presents a new method for classifying malware algorithms that exhibits both high accuracy and high coverage. The method combines big data analysis with software security technologies such as feature extraction, machine learning, binary instrumentation and dynamic instruction flow analysis to achieve automated classification of malware algorithms. 20 classification experiments prove the correctness of the method. We also discuss future directions for improving the method.

关键词： malware analysis algorithm classification big data feature extraction machine learning binary instrumentation dynamic instruction flow

来源：评论

学校读者我要写书评

暂无评论

Research on intelligent classification and grade regression prediction of metal ores in complex marine environments

引用

MARINE GEORESOURCES & GEOTECHNOLOGY 2025年

作者： Wang, Jianhua Yin, Tubing Dai, Hao Ma, Jiexin Cent South Univ Sch Resources & Secur Engn Changsha Peoples R China Shenzhen Port Grp Co Ltd Shenzhen Peoples R China

To address the high cost and complexity of marine exploration technology, this study proposes a novel approach that utilizes an underwater robot equipped with a metal detector to identify ore types and grades on the seabed directly. By integrating laboratory physical experiments, computer vision technology, classification algorithms, and mathematical regression analysis, multiple types of metal ores are classified, and the grade of major minerals is predicted under the signal attenuation effects of PVC boards and seawater. The key findings are as follows: improved Random Forest, Logistic Regression, AdaBoost, and SVC models were employed to classify the ore spectral images obtained from laboratory experiments. Among the four models, the Random Forest model had the lowest test accuracy, approximately 0.91, while both the Logistic Regression and AdaBoost models achieved slightly higher accuracies, around 0.94. The SVC model demonstrated the best performance, with the highest test accuracy, close to 0.97. Additionally, as the thickness of the PVC board increased, the grade of the primary elements in the test samples exhibited an exponential decline, with a rapid initial attenuation. However, with further increases in PVC board thickness, the downward trend of mineral grade became more gradual, and the rate of change decreased.

关键词： Marine exploration Mineral grade algorithm classification Support vector machine regression prediction

来源：评论

学校读者我要写书评

暂无评论

A Hybrid Machine Learning Model for Code Optimization

引用

INTERNATIONAL JOURNAL OF PARALLEL PROGRAMMING 2023年第6期51卷 309-331页

作者： Hakimi, Yacine Baghdadi, Riyadh Challal, Yacine Ecole Natl Super Informat Lab Methodes Concept Syst LMCS BP 68M Oued Smar 16309 Algeria MIT Cambridge MA USA New York Univ Abu Dhabi NYUAD Abu Dhabi U Arab Emirates Univ Doha Sci & Technol UDST Coll Comp & Informat Technol Doha Qatar

The complexity of programming modern heterogeneous systems raises huge challenges. Over the past two decades, researchers have aimed to alleviate these difficulties by employing classical Machine Learning and Deep Learning techniques within compilers to optimize code automatically. This work presents a novel approach to optimize code using at the same time Classical Machine Learning and Deep Learning techniques by maximizing their benefits while mitigating their drawbacks. Our proposed model extracts features from the code using Deep Learning and then applies Classical Machine Learning to map these features to specific outputs for various tasks. The effectiveness of our model is evaluated on three downstream tasks: device mapping, optimal thread coarsening, and algorithm classification. Our experimental results demonstrate that our model outperforms previous models in device mapping with an average accuracy of 91.60% on two datasets and in optimal thread coarsening task where we are the first to achieve a positive speedup on all four platforms while achieving a comparable result of 91.48% in the algorithm classification task. Notably, our approach yields better results even with a small dataset without requiring a pre-training phase or a complex code representation, offering the advantage of reducing training time and data volume requirements.

关键词： Machine learning Deep learning Code optimizations LLVM-IR Heterogeneous platforms Thread coarsening algorithm classification Code mapping

来源：评论

学校读者我要写书评

暂无评论

A Game-Based Framework to Compare Program Classifiers and Evaders 2023

A Game-Based Framework to Compare Program Classifiers and Ev...

引用

21st ACM/IEEE International Symposium on Code Generation and Optimization (CGO)

作者： Damasio, Thais Canesche, Michael Pacheco, Vinicius Botacin, Marcus da Silva, Anderson Faustino Quintao Pereira, Fernando M. Univ Fed Minas Gerais Belo Horizonte MG Brazil Texas A&M Univ College Stn TX 77843 USA Univ Estadual Maringa Maringa Parana Brazil

ISBN: (纸本)9798400701016

algorithm classification consists in determining which algorithm a program implements, given a finite set of candidates. Classifiers are used in applications such malware identification and plagiarism detection. There exist many ways to implement classifiers. There are also many ways to implement evaders to deceive the classifiers. This paper analyzes the state-of-the-art classification and evasion techniques. To organize this analysis, this paper brings forward a system of four games that matches classifiers and evaders. Games vary according to the amount of information that is given to each player. This setup lets us analyze a space formed by the combination of nine program encodings;seven obfuscation passes;and six stochastic classification models. Observations from this study include: (i) we could not measure substantial advantages of recent vector-based program representations over simple histograms of opcodes;(ii) deep neural networks recently proposed for program classification are no better than random forests;(iii) program optimizations are almost as effective as classic obfuscation techniques to evade classifiers;(iv) off-the-shelf code optimizations can completely remove the evasion power of naive obfuscators;(v) control-flow flattening and bogus-control flow tend to resist the normalizing power of code optimizations.

关键词： algorithm classification obfuscation

来源：评论

学校读者我要写书评

暂无评论

Complexity-Based Code Embeddings 15th

Complexity-Based Code Embeddings

引用

15th International Conference on Computational Collective Intelligence (ICCCI)

作者： Folea, Rares Iacob, Radu Slusanschi, Emil Rebedea, Traian Univ Politehn Bucuresti Bucharest Romania

ISBN: (纸本)9783031414558;9783031414565

This paper presents a generic method for transforming the source code of various algorithms to numerical embeddings, by dynamically analysing the behaviour of computer programs against different inputs and by tailoring multiple generic complexity functions for the analysed metrics. The used algorithms embeddings are based on r-Complexity [7]. Using the proposed code embeddings, we present an implementation of the XGBoost algorithm that achieves an average 90% F1-score on a multi-label dataset with 11 classes, built using real-world code snippets submitted for programming competitions on the Code-forces platform.

关键词： algorithm classification code embeddings complexity tree-based classification

来源：评论

学校读者我要写书评

暂无评论

From driving trajectories to driving paths: a survey on map-matching algorithms

引用

CCF TRANSACTIONS ON PERVASIVE COMPUTING AND INTERACTION 2022年第3期4卷 252-267页

作者： Jiang, Linli Chen, Chaoxiong Chen, Chao Huang, Hongyu Guo, Bin Chongqing Univ Coll Comp Sci Chongqing Peoples R China Northwestern Polytech Univ Sch Comp Sci Xian Peoples R China

With the widespread deployment of built-in Global Positioning System (GPS) devices, numerous volumes of driving trajectories can be recorded conveniently. Affected by GPS equipment precision and driving environments, raw GPS trajectories will deviate from the paths that vehicles really drove on. Such inaccurate data is not beneficial to the upstream applications. Therefore, map matching is applied to identify the true driving paths in the road network from the GPS trajectories data. A lot of studies in the filed of map matching have been proposed, but there still exist three problems: (1) there lacks a comprehensive review on recent map matching algorithms with new techniques;(2) the existing map-matching algorithms still fail to meet the requirements of both high precision and high efficiency simultaneously;(3) there is a lack of comparison between various types of matching algorithms on a unified experimental environment. In this paper, we review the existing matching algorithms and propose a new categorisation based on their methodologies. The proposed categorisation can better reveal their properties and facilitate the future utilization. In addition, we conduct an experimental comparison among four representative algorithms to give a deep insight to the properties of different categories. Experimental results reveal the importance of some solutions to improve matching accuracy and efficiency.

关键词： Map matching algorithm classification GPS trajectory Survey

来源：评论

学校读者我要写书评

暂无评论

Map-Matching Techniques for Train Localization: A Taxonomic Survey

引用

IEEE ACCESS 2024年 12卷 192328-192340页

作者： Millan-Jimenez, Iker Zabalegui, Paul de Miguel, Gorka Mendizabal, Jaizki Marcos, Inigo Adin CEIT Basque Res & Technol Alliance BRTA Res Ctr Donostia San Sebastian 20018 Spain Univ Navarra TECNUN Donostia San Sebastian 20018 Spain

Precise localisation of rail vehicles is a key element towards the development and deployment of novel train control systems. Localisation of trains can be considered as a 1-D localisation problem, as trains only move on tracks, easing the localisation approach. Localisation with maps makes a great choice as the maps describe unambiguously train tracks, easing the train localisation and making it an effective choice. On top of that, train localisation with a map requires just onboard sensors making it a low-cost alternative to localisation with track-side equipment, as this equipment makes train localisation costs increase with the reach of the use. Even though this topic has been widely studied in the literature, there has not been a classification of the used methodologies for map-matching for train localisation. Therefore, the purpose of this paper is to make a classification of the state-of-the-art map-matching algorithms for train localisation. On top of that, this paper also discusses how to build a digital map from a set of coordinates. Three main categories have been identified in the literature for map-matching: geometric, similarity and hypothesis. And mainly three types of digital maps have been observed in the literature: interpolation, splines and geometric. This paper helps practitioners and researchers have a comprehensive foundation on map-matching for train localisation.

关键词： Surveys Rail transportation Sensors Geometry classification algorithms Accuracy Databases Tracking Global navigation satellite system Splines (mathematics) algorithm classification digital map mapmatching train localization

来源：评论

学校读者我要写书评

暂无评论

Resource Management Techniques for Cloud/Fog and Edge Computing: An Evaluation Framework and classification

引用

SENSORS 2021年第5期21卷 1832页

作者： Mijuskovic, Adriana Chiumento, Alessandro Bemthuis, Rob Aldea, Adina Havinga, Paul Univ Twente Dept Pervas Syst NL-7522 NB Enschede Netherlands Univ Twente Dept Ind Engn & Business Informat Syst NL-7522 NB Enschede Netherlands

Processing IoT applications directly in the cloud may not be the most efficient solution for each IoT scenario, especially for time-sensitive applications. A promising alternative is to use fog and edge computing, which address the issue of managing the large data bandwidth needed by end devices. These paradigms impose to process the large amounts of generated data close to the data sources rather than in the cloud. One of the considerations of cloud-based IoT environments is resource management, which typically revolves around resource allocation, workload balance, resource provisioning, task scheduling, and QoS to achieve performance improvements. In this paper, we review resource management techniques that can be applied for cloud, fog, and edge computing. The goal of this review is to provide an evaluation framework of metrics for resource management algorithms aiming at the cloud/fog and edge environments. To this end, we first address research challenges on resource management techniques in that domain. Consequently, we classify current research contributions to support in conducting an evaluation framework. One of the main contributions is an overview and analysis of research papers addressing resource management techniques. Concluding, this review highlights opportunities of using resource management techniques within the cloud/fog/edge paradigm. This practice is still at early development and barriers need to be overcome.

关键词： resource management cloud computing fog computing edge computing algorithm classification evaluation framework

来源：评论

学校读者我要写书评

暂无评论

AutoFocus: Interpreting Attention-based Neural Networks by Code Perturbation 34

AutoFocus: Interpreting Attention-based Neural Networks by C...

引用

34th IEEE/ACM International Conference on Automated Software Engineering (ASE)

作者： Bui, Nghi D. Q. Yu, Yijun Jiang, Lingxiao Singapore Management Univ Sch Informat Syst Singapore Singapore Open Univ Sch Comp & Commun Milton Keynes Bucks England

ISBN: (纸本)9781728125084

Despite being adopted in software engineering tasks, deep neural networks are treated mostly as a black box due to the difficulty in interpreting how the networks infer the outputs from the inputs. To address this problem, we propose AutoFocus, an automated approach for rating and visualizing the importance of input elements based on their effects on the outputs of the networks. The approach is built on our hypotheses that (1) attention mechanisms incorporated into neural networks can generate discriminative scores for various input elements and (2) the discriminative scores reflect the effects of input elements on the outputs of the networks. This paper verifies the hypotheses by applying AutoFocus on the task of algorithm classification (i.e., given a program source code as input, determine the algorithm implemented by the program). AutoFocus identifies and perturbs code elements in a program systematically, and quantifies the effects of the perturbed elements on the network's classification results. Based on evaluation on more than 1000 programs for 10 different sorting algorithms, we observe that the attention scores are highly correlated to the effects of the perturbed code elements. Such a correlation provides a strong basis for the uses of attention scores to interpret the relations between code elements and the algorithm classification results of a neural network, and we believe that visualizing code elements in an input program ranked according to their attention scores can facilitate faster program comprehension with reduced code.

关键词： attention mechanisms neural networks algorithm classification interpretability explainability code perturbation program comprehension

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：